PROJECT SUMMARY/ABSTRACT Alzheimer's disease (AD) is the most costly and burdensome disease in the U.S. Its public health impact will only grow with the increase of 65-75 year olds in the next decade. The AD process begins 2 or more decades before dementia onset. Identifying individuals during early stages (e.g., mild cognitive impairment ([MCI]) is estimated to result in massive savings. Thus, NIH and Alzheimer's Association consensus statements emphasize early identification. Like cardiovascular disease, focusing on middle age is crucial for earlier identification of risk for cognitive decline, preclinical AD, and MCI. Despite this protracted progression of AD pathology, little is known about its temporal course and relation to cognition in middle age. To address this critical knowledge gap, we propose to collect a fourth wave of data in the Vietnam Era Twin Study of Aging (VETSA). VETSA provides detailed characterization of change throughout midlife. With wave 4, VETSA will cover an 18-year period in our community-dwelling sample. VETSA began with virtually all subjects in their 50s, so we can track shifts from normal cognition to MCI/AD and normal to abnormal biomarker status. We focus on 4 sets of indicators that improve the ability to identify at-risk individuals earlier than in most studies: 1) extensive cognitive testing; 2) plasma AD biomarkers for beta-amyloid (A?), tau, and neurofilament light (NfL); 3) polygenic risk scores; and 4) novel assessments of cognitive processes. Almost all subjects will have been A?- at VETSA 1. Average age at VETSA 4 will be 74, and a meta-analysis indicates that >30% of non- demented adults at age 75 are A?+. Thus, leveraging data from previous waves, the timing is ideal to capture the transition to disease states. We utilize the amyloid-tau-neurodegeneration (ATN) biomarker classification system and examine the proposed A!T!N staging of the AD continuum. Naturally, most research focuses on ATN biomarkers as predictors, but it would be highly advantageous to identify people at risk before reaching pathological A? levels. Thus, Aim1 will examine plasma ATN biomarker trajectories as well as predictors of ATN biomarker accumulation and abnormality. We have biomarker data from VETSA 1 and 3, and will perform assays on VETSA 4 data. Aim 2 models risk and protective factors for cognitive decline, biomarker trajectories, and progression to MCI using genetically-informative analyses that can test causality. With 4 time points, we will use our combination of twin and polygenic risk score data and our extensive health/medical and psychosocial data to elucidate factors accounting for accelerated cognitive decline. In Aim 3, we evaluate 2 novel early risk indicators by: a) extending our work on pupil dilation as a measure of cognitive effort before cognitive performance declines; and b) assessing visual short-term memory binding, an early indicator tested primarily in familial AD families. To increase detection of decline, Aim 4 adds telephone/mailed assessments partway in the funding period. Wave 4 will have N=1000. VETSA's unique features make it a most promising resource for advancing knowledge about early identification, with potential for a profound public health impact.